INDEX

Explanations

nouns with high frequency and importance in the document context

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Schmidt

-0.15

cks

-0.15

auer

-0.15

ovol

-0.14

 Hour

-0.14

 Balt

-0.14

atak

-0.13

ctl

-0.13

Äĥn

-0.13

azz

-0.13

POSITIVE LOGITS

.scalablytyped

0.17

éĳĳ

0.15

oved

0.15

ume

0.14

]int

0.14

iu

0.14

Persistence

0.14

une

0.14

isco

0.13

amespace

0.13

Activations Density 0.038%

nouns with high frequency and importance in the document context

No Comments

No Known Activations